Development of a Distributed Energy Router Control System Based on a Neural Network

Abstract

The article describes the implementation of the “Energy Internet” concept at the 6–20 kV medium-voltage level. Energy router is a basic power flow control device in a distribution electric network (DEN). A decentralized energy router control system based on an artificial neural network has been proposed. The paper demonstrates the stages of its training and testing. A DEN model with an energy router has been created. The parameter dataset of the DEN model has been obtained by calculation. The data have been filtered and optimized on the developed algorithm. The generated dataset was used for training and testing the control system in a computer DEN model. The article shows the efficiency of using an energy router controlled by the developed algorithm.